Origin of Sedimentary Cycles in Mixed Carbonate–Siliciclastic Systems

Abstract

The separation of eustatic, tectonic, and other controls on the development of sedimentary cyclicity is difficult. In mixed carbonate-siliciclastic successions, the conventional interpretation of unconformity-bounded depositional sequences is that they are due to reciprocal sedimentation in response to relative changes of sea level. According to this view, transgressive and highstand systems tracts are composed primarily of carbonate rocks, and lowstands of siliciclastic rocks. The application of this model to the interpretation of cyclic carbonate and siliciclastic rocks in the Upper Devonian of the Canning basin, Western Australia, presents a paradox because expected evidence for subaerial exposure of the platform is not well developed. Sequence stratigraphic studies in outcrop at two localities along the northern margin of the Canning basin confirm a complex relation between carbonate and siliciclastic conglomerate. At Stony Creek, the carbonate rocks are interpreted to represent an assemblage of reef, foreslope floatstones, and backreef carbonate-conglomerate cycles that accumulated in a shallow marine environment along the margin of a fan-delta. Conglomerates and sandstones inferred to onlap the reefal foreslope are interpreted tentatively as fluvial, and the contact is interpreted as a sequence boundary. In the Van Emmerick Range, foreslope floatstones are onlapped along the margin of an incised valley with at least 10 m of relief by conglomerate and sandstone of probable marine origin, and overlain by a transgre sive fossiliferous limestone. The age of the sequence boundary at Stony Creek is not well established, but probably early Frasnian. The age of the sequence boundary in the Van Emmerick Range is better constrained as late Frasnian, and this surface appears to correlate with the base of the Frasnian 4 sequence identified by Southgate et al. (1993) on the basis of subsurface data. Both surfaces are thought to have involved base-level lowering, but the evidence is equivocal. End_Page 181------------------------ Evaluation of available data indicates that subaerial exposure is required for only one of eight potential sequence boundaries in the Frasnian-Famennian interval, a surface that corresponds locally to minor karstification, and which is dated as latest Frasnian. Several explanations are proposed as working hypotheses. The development of thick lowstand deposits coeval with flooding events on the platform is consistent with continued extension and tilting of a fault block. Alternatively, exposure may have been restricted to topographic highs remaining after extension had ceased. If subaerial exposure was widespread, diagenetic effects may have been limited or not preserved. The development of onlap surfaces within the basin may be related in part to variations in sediment flux, and the d stribution of siliciclastic sediments influenced by the geologic structure, especially the configuration of accommodation or transfer zones. Further work is needed to resolve uncertainties in the existing sequence stratigraphic interpretation, to improve the calibration of individual boundaries, and to evaluate these ideas. Ultimately, comparisons with coeval successions on other continents will be needed to evaluate the possible role of eustasy in the development of the observed sequences.